Population and Security:

How Demographic Change Can

Lead to Violent Conflict

JACK A. GOLDSTONE
_________________________

While the marked decrease in population growth in many countries and

regions is good news for those concerned about global popula tion, it
offers no clear relief for concerns about the security implications
of population change.
_________________________

A s we cross into the new century, the world seems finally to

have turned the corner on population growth. A
combination of increased education for women, national and
international support for policies of population planning and the
spread of economic development and accompanying movement
along the demographic transition frontier have led to falling
population growth rates around the world. Whether among the
behemothsChina and Indiaor among the smaller but rapidly
growing nationssuch as Saudi Arabia, Kenya and Malawi
population growth rates have dropped dramatically in the last
decade.i
Yet while population growth rates have dropped around the
world, they remain high in some areas. In particular, many
nations in the Middle East, southeast Asia and central and
northern Africa still are growing at nearly 3 percent per year, a
growth rate that leads to a doubling of population in
approximately 25 years. Moreover, although in most countries
the rate of population growth has slowed, the absolute number
of people being added to the worlds population has not; the

_____________________________
Journal of International Affairs, Fall 2002, vol. 56, no. 1. The Trustees of
Columbia University in the City of New York.
 Population and Security

large number of women of childbearing age in the developing

world, carrying the momentum of past population growth,
ensure that even while growth rates fall as a percentage of the
existing population, the number of new births each year
continues to rise. For example, although Chinas growth rate has
fallen to 1.0 percent per year, China will still grow by 10 to 11
million people per year for the next 15 years. The world as a
whole will add roughly 80 million people per year, or another
960 million (that is, another India) in the next dozen years.ii

DEMOGRAPHY AND SECURITY: KEY FINDINGS

After nearly three decades of debate and analysis, stemming
from Myron Weiners (1971) path-breaking study, scholars are
beginning to develop much clearer answers to the complex
questions regarding how population changes affect security
concerns. Those answers can be summarized briefly in the
following propositions, each of which we shall treat in greater
detail below:
1) While population growth often brings degradation of
forests, water resources, arable land and other local resources,
such environmental degradation is not a major or pervasive
cause of international wars, ethnic wars or revolutionary
conflicts. Such degradation often brings misery, yet such misery
does not generally trigger the elite alienation and opposition to
the government necessary for large-scale violence to occur.
2) Population growth can give rise to conflicts over
increasingly scarce resources, such as farmland, if those conflicts
involve elites seeking to take resources from popular groups, or
competition between elite factions for control of those resources.
However, what determines whether violent conflict arises are the
relationships among popular groups, elites and the state, and
particularly whether the state has the capacity to channel and
moderate elite conflicts. Only where elite conflicts or popular
resistance to elite actions overwhelm weak states do major
conflicts arise.
3) While overall population growth and population density
do not generally predict political risks, a number of distinct

FALL 2002 | 4
 Population and Security

kinds of demographic changesrapid growth in the labor force

in slow-growing economies, a rapid increase in educated youth
aspiring to elite positions when such positions are scarce,
unequal population growth rates between different ethnic
groups, urbanization that exceeds employment growth and
migrations that change the local balance among major ethnic
groupsdo appear to increase the risks of violent internal
political and et hnic conflicts. In addition, there is some evidence
that countries with larger populations have greater risks of both
armed conflict and state repression.
4) Most population changes do not directly increase the risks
of international wars between domestically stable states;
however, because many international wars have their origins in
domestic conflicts (e.g., the Iran/Iraq war growing out of Irans
revolution; international wars in West and Central Africa
growing out of the collapses of Liberia, Sierra Leone and
Congo/Zaire), in those contexts where population changes
produce domestic political crises, the risk of international war is
also increased. There is also some evidence that the intensity of
war, in terms of casualties, increases in countries with
exceptionally large youth cohorts.
5) Certain demographic changes, such as a rise in infant
mortalityaside from whatever role they may have as causes
can be powerful indicators of coming political violence.
6) Rapid and large-scale demographic changes, such as a rise
in mortality or a sharp rise in migration, can arise as an outcome
of violent conflicts.

THE ENVIRONMENT AS A CAUSE OF VIOLENT CONFLICTS

Thomas Homer-Dixon provoked a great deal of controversy and
concern with his claim that we are on the threshold of an era
in which armed conflicts will arise with increasing frequency as a
result of environmental change.iii However, in the years since his
warning, the search for evidence behind this claim has provided
little support. As Paul Diehl has remarked, the many
publications from the [Toronto] project have produced largely
abstract conceptions of the environment-conflict nexus, with

FALL 2002 | 5
 Population and Security

actual cases presented only as anecdotal evidence or as

illustrative examples.iv After nearly a decade of research, it now
seems clear that long-term environmental degradation of the
kind that often accompanies development (e.g., soil erosion,
deforestation and air and water pollution) has little or no
significant role in generating civil or international wars.v
Detailed cross-national studies have found only very weak
relations between environmental degradation and either
international or domestic armed conflict.vi In most studies that
make an effort to measure the relative impact of environmental
and other causes, environmental factors emerge as less
important in determining the incidence of civil conflict than
economic and political factors.vii
For example, Wenche Hauge and Tanja Ellingsen, in the
most comprehensive global test of the environmental-scarcity-
leads-to-violence hypothesis with recent data (198092), found
that while deforestation, land degradation and low freshwater
availability were positively correlated with the incidence of civil
war and armed conflict, the magnitude of their effects was tiny.
By themselves, these factors raised the probability of civil war by
0.5 to under 1.5 percent.viii These factors did have a slightly
higher impact on the probability of lesser kinds of armed conflict
(causing increases in the chances of such conflict by from 4
percent to 8 percent); but their influence paled compared to the
impact of such traditional risk factors as poverty, regime type
and current and prior political instability.
In addition, Gnther Baechlers extensive study of the
relationships between environmental change and violent conflict
found that while environmental degradation could be a
background or triggering factor in ethnic or political conflicts,
most such conflicts were local and peacefully resolved by
government regulation or negotiations. Whether or not such
conflicts pass[ed] the threshold of violence definitely depends
on sociopolitical factors and not on the degree of environmental
degradation as such.ix
A third study, undertaken by an academic Task Force on
State Failure sponsored by the US government,x deliberately

FALL 2002 | 6
 Population and Security

sought environmental causes for a wide range of violent conflict

events, including authoritarian coups, revolutionary wars, ethnic
wars and genocides. However, after adjusting for the impact of
living standards, regime type and involvement in international
trade, no direct impact of environmental variables could be
found.
It must be admitted that the range and quality of data on
environmental change leaves much to be desired, and the
poverty of such data may be one reason for these negative
findings. Still, if environmental change were truly a major and
pervasive cause of violent conflicts, it seems likely that some
large cross-national studies of recent political violence would
show more positive findings.
Should we therefore dismiss the environment as a cause of
conflict? No, although I believe we can be free of the fear that
environmental decay will unleash wars and revolutions across
the globe. Rather, what research has shown is that although
environmental issues do cause international and domestic
conflicts, they are of the kind that are generally settled by
negotiation and compromise and do not lead to taking up arms.
The reason for that is straightforward. Where the problem
faced by two groups, or two nations, is over the degradation or
depletion of an environmental resource, war neither solves the
problem (it cannot make more of the resource) nor is it an
economically efficient way to redistribute the resource (the costs
of war almost invariably far outweigh the cost of gaining
alternative resources or paying more for a share of the resource).
For example, if two nations have a conflict over sharing river
watersuch as India and Bangladesh over the Ganges,xi Israel
and Jordan over the river Jordanxii or Hungary and Slovakia over
the Danubexiiithey may threaten violence but in fact are most
likely to produce non-violent resolution through negotiation or
arbitration rather than war (and indeed all of these conflicts led
to treaties or international arbitrationxiv). The reason is that for
one party to insist on all the water would in fact be a casus belli;
and to risk a war to simply increase ones access to water is
economically foolhardy. Throughout the world, the main use of

FALL 2002 | 7
 Population and Security

freshwater (over three-quarters) is for irrigation to produce food.

A reduction in water can be compensated either by adopting
more efficient means of irrigation (drip rather than ditch); by
switching to less water-intensive crops (dry grains rather than
rice; tree crops rather than grains); or by importing food rather
than producing it. All of these steps, though costly, are far, far,
less costly than armed conflict. Thus for both the country with
the ability to take more water and the country dependent on
downstream flows, the issue will be how to use and negotiate use
of the resource most efficiently; resort to war would inevitably
be more costly than any gains that could be made from
increased access to the resource. No nations have ever gone to
war strictly over access to water; nor are any likely to do so in
the future.xv

ELITES AND VIOLENT CONFLICTS

Much of the literature on environmental scarcity and violent
conflict has erred in predicting violence because of a
fundamental misunderstanding regarding the causes of political
crises. It is a profound and repeated finding that the mere facts
of poverty and inequality or even increases in these conditions,
do not lead to political or ethnic violence.xvi In order for popular
discontent or distress to create large-scale conflicts, there must
be some elite leadership to mobilize popular groups and to
create linkages between them. There must also be some
vulnerability of the state, in the form of internal divisions and
economic or political reverses. Otherwise, popular discontent is
unvoiced, and popular opposition is simply suppressed.
Political analysts of violent conflict now recognize that the
essence of political stability or instability lies in a set of
reciprocal relationships: among states in the international
system, between states and their societys elites, among elite
factions and between both states and elites and popular groups.
When states are fiscally sound, free of severe international
threats and supported by their elites, they are enormously
resistant to popular discontent. It is only when states become
financially strapped or subject to international pressure, and are

FALL 2002 | 8
 Population and Security

deserted by their elites, that popular distress furnishes raw

material for mobilizing forces for conflict.xvii
Where land scarcity or other resource shortages appear to
play a role in violent conflictsin South Africa or Kenya, for
examplexviiithe essence of the conflict has generally been the
struggle among elite factions for control of political power, with
control of land simply representing one of the prizes that go to
the winning faction. Without political struggles that turn elites
against the state, or that turn elite factions against each other,
large-scale political conflicts are simply unlikely to arise. While
the control of landlike the control of mineral or other
resourcesmay figure in such struggles, the degradation of
environmental resources is generally is not a significant enough
factor to be a major cause of violent conflicts.

POPULATION CHANGES AND VIOLENT CONFLICTS

It is true that overall population growth, or increases in
overall population density, do not generally lead to
violent conflict. But research has shown a variety of
instances in which particular kinds of population
changes are strongly associated with political
instability.xix
Alex de Sherbinens comment points out that viewing the
impact of population change on conflict only in terms of overall
population growth is too simple to capture the complex
relationships involved. Rather, it is particular kinds of
demographic changes, occurring in particular political and
economic contexts, that cause instability.
For example, if an agrarian population that needs more land
to provide for a growing population finds that adjacent land is
owned, and even being expanded, for exclusive use by large
landowners, conflict is likely and indeed nearly inevitable.
Throughout history, confrontations over land between growing
populations of peasants and large landholders have prompted
rural rebellions in China, Latin America and Europe. In most
such cases, there is no environmental degradationpeasants and
landowners alike are often improving the land. However,

FALL 2002 | 9
 Population and Security

population growth leads to the cultivation of more marginal

lands and incursions by land-hungry peasants into areas also
sought by profit-hungry landlords. The result is a combination
of pressure on peasant incomes and heightened conflicts with
local elites. Conflict of this sort has arisen most recently in
Chiapas in Mexicoxx but is typical of peasant/landlord relations
throughout history, appearing in the French Revolution of 1789,
the German Revolution of 1848, the Mexican Revolution of
1910, the Russian Revolution of 1917 and the Chinese
Revolution of 1949.xxi
Such rural conflict can be avoided if the urban and industrial
economy provides sufficient jobs to absorb an expanding
population. However, studies have shown that where urban
growth is not matched by an increase in economic growth, risks
of political turbulence increase.xxii A recent study of political
crises in sub-Saharan Africa from 1955 to 1995 by the State
Failure Task Force found that, other things equal, the risk of
political crisis nearly doubled in countries with above-average
levels of urbanization but below-average levels of GDP/capita.xxiii
The problem of over-urbanization relative to incomes is just
one aspect of a more general principle relating population
changes to political instability, namely that problems arise when
there is a persistent mismatch between employment prospects
and the size and nature of the labor force. Thus not only over-
urbanization, but also over-education relative to the caliber of
available jobs can create political discontent. In revolutionary
situations ranging from Tudor England to Enlightenment
France, from late Tokugawa Japan to modern Iran and the
Soviet Union, political upheaval has been preceded by a surge in
the production of youth with advanced education in the context
of a relatively limited, semi-closed structure of elite positions.xxiv
The central authorities, who guarded the gates of social and
economic advancement, drew elite discontent for a situation in
which social mobility was increa singly sought but the paths of
mobility were increasingly clogged.
Even without increases in higher education, the rapid growth
of youth can undermine existing political coalitions, creating

FALL 2002 | 10
 Population and Security

instability. Large youth cohorts are often drawn to new ideas

and heterodox religions, challenging older forms of authority.xxv
In addition, because most young people have fewer
responsibilities for families and careers, they are relatively easily
mobilized for social or political conflicts.xxvi Youth have played a
prominent role in political violence throughout recorded history,
and the existence of a youth bulge (an unusually high
proportion of youths 15 to 24 relative to the total adult
population) has historically been associated with times of
political crisis. Most major revolutionsthe English Revolution
of the seventeenth century, the French revolution of the
eighteenth century and most twentieth-century revolutions in
developing countrieshave occurred where exceptionally large
youth bulges were present.xxvii
Christian Mesquida and Neil Wiener have presented data
showing that the severity of conflicts, as measured by the
number of deaths in armed conflict, is much higher for countries
that have a large youth bulge, even when controlling for the
effects of income and inequality.xxviii Henrik Urdal has analyzed
the effects of youth bulges on a wide variety of conflicts.
Although he finds that youth bulges are not significantly
associated with civil wars, they do appear to be associated with
the onset of smaller violent conflicts, involving less than 1,000
deaths. Moreover, they seem to have an increasingly large effect
as the youth bulge grows more extreme, and particularly in
transitional regimes with neither a fully democratic nor fully
autocratic character.xxix
A number of researchers have also produced results showing
that the size of popula tion itself influences both the incidence of
conflict, and the degree of state repression.xxx Their argument is
that larger populations both require more intense state action to
suppress dissent and offer more opportunities for opposition
groups to recruit and mobilize supporters. There are also
generally more opportunities for clashes between different
regional or ethnic groups, other things being equal, in larger
populations. Some research has also found that state repression
is higher in more rapidly growing populations.xxxi However, other

FALL 2002 | 11
 Population and Security

scholars have found different results. The State Failure Task

Force, for example, has so far found no significant effects for
population size, population density, population growth or even
youth bulges as a cause of violent conflicts.
These various results suggest a cautious approach to
determining the precise effect of population variables on
conflict. Part of the reason for the varying results may be that
different researchers have relied on different methods to analyze
their data. In fact, there is considerable controversy over how to
analyze data sets made up of large numbers of 0/1 (peace/war)
observations from many counties across many years, especially
when the observations of war (1s) are relatively few compared to
observed years of peace (0s). Conventional approaches to correct
for auto-correlation run into difficulty compensating for
simultaneous problems of spatial auto-correlation, temporal
auto-correlation and rare-events bias; scholars have not yet
determined which method will give the best estimates of the
effects of population and other variables in such data.xxxii
Different approaches give different estimates of the significance
of particular variables.
However, a more important reason for disagreements is that
demographic factors are only a part of the complex causal forces
behind violent conflicts. It is generally agreed that economic
development (as measured by GNP/capita or infant mortality)
and regime type (autocracy, democracy or
transitional/intermediate) are critical elements influencing the
risks of conflict. A host of other factorsleadership, colonial
experience, terrain, trade and state discriminationhave also
been suggested as playing a role. Population factors may interact
and overlap with other such causal forces in varying degrees
For example, the State Failure Task Force has found that a
countrys involvement in international trade seems to have a
moderating effect on conflict---risks of political crises are lower,
other things being equal, for countries whose imports and
exports are large compared to its gross domestic product.xxxiii It
appears that extensive involvement in international trade is
either an indicator or cause of fairly stable economic

FALL 2002 | 12
 Population and Security

relationships among elites or perhaps a source of income for the

regime that restrains conflict. However, countries with larger
populations also generally provide more of their own food and
manufactures, while smaller countries depend more on exports
and imports to meet the full range of their investment and
consumption needs. Thus, larger populations correlate with
lower trade involvement. The task force finds that both larger
population size and lower trade involvement are associated with
higher risks of conflict and that when both variables are included
in models of conflict processes, the impact of trade involvement
generally remains statistically significant, while that of
population size does not. It may be that population size affects
conflict mainly through its relation to trade openness; but most
of the other researchers who have found that larger populations
have higher incidence of violent conflict have not included trade
variables in their models.
It is thus too early to treat the preceding results on the
impact of demographic factors on violence as definitive.
Problems of method and of testing complex and varied models
to explore various constellations of demographic and other
causes, still remain.xxxiv We might find that the relationships of
demographic factors to conflict do not even fit the traditional
notion of necessary and sufficient conditions as the basis for
causal relationships.xxxv The historical case data makes it appear
that youth bulges might well be necessary, even if they are not
sufficient, for the occurrence of large-scale violent conflicts.
Population movements across, or even within, political
borders can also lead to violence. The US Indian Wars of the
eighteenth and nineteenth centuries were caused by the
expansion of the United States into already-settled Native
American territ ories. The state-assisted migration of Han
Chinese into the mainly Uighur-settled region of Xinjiang and
into Tibet has led to violent episodes of rebellion in both
regions, as their inhabitants struggled to maintain their
distinctive identities and control over their territories. The
Bantu migrations into southern Africa led to wars throughout
the continent, while the movement of peoples, both forced and

FALL 2002 | 13
 Population and Security

by choice, across ethnic borders within the former Soviet Union

has led to a legacy of ethnic and separatist conflicts.xxxvi
The crucial element here is not migration per se; economic
migration often leads to substantial benefits for both migrants
and the destination country. What appears to matter for conflict
are those cases wherein migration leads to clashes of national
identity.xxxvii That is, when one distinct ethnic group migrates
into an area that is considered homeland by another ethnic
group and challenges the dominance of the latter, then conflicts
are likely to arise. If these conflicts escalate into contests for
political control of the region, then ethnic war and even
genocide often results.
To sum up, the still incomplete but growing body of
evidence and analysis that we have argues that a number of
specific population changes are strongly associated with
increased risks of political violence:
1) An expanding agrarian population running up against
land that is controlled or being expanded for exclusive use of
large landlords;
2) An expanding urban population in an economy that is not
providing commensurate economic growth;
3) An expanding population of higher-educated youth facing
limited opportunities to obtain elite political and economic
positions;
4) A large youth bulge; that is, an expansion of the 15 to 25
age cohort relative to the overall adult population of a society,
especially where political institutions are weak;
5) The migration of populations into regions already settled
by a population with a distinct ethnic or political identity.
Clearly, none of these conditions arises from population
growth or even from specific population changes by themselves.
The conditions that lead to violent conflicts involve population
changes in specific contexts where there are blockages to the
desires or needs of an expanding population. Thus, if we wish to
know in what regions of the globe we are most likely to see
population-induced political conflicts, we need to examine both

FALL 2002 | 14
 Population and Security

expected population changes and the contexts in which they will

occur.

FUTURE POPULATION CHANGES AND RISKS OF VIOLENT

CONFLICT
Even countries with relatively low growth rates may encounter
situations in which population changes contribute to political
violence: It is not the absolute rate of population growth but the
imbalance between growth in specific sectors of the population
and growth of the economy that is crucial to the creation of
conflicts. For example, from 1970 to 1991 in the USSR, when
economic growth slowed almost to zero, population growth was
also minimal. However, the Soviet Union still encountered four
of the five demographic risk conditions noted above, namely
2)an urban population that continued to grow despite
minimal economic growth; 3)an over-expansion of young men
with a technical higher education, most of whom were relegated
to blue-collar jobs due to party restrictions on entry to the
managerial and political elites and a stagnant economy; 4)a
large youth bulge in the Central Asian republics; and 5)large-
scale migration of Russians into many non-Russian ethnic soviet
republics. All of these factors became important in mobilizing
the urban and nationalist oppositions whose combination
produced the collapse of the Communist regime.xxxviii
It is precisely because of the importance of such imbalances
that countries such as Saudi Arabia and China bear watching for
political unrest, despite their success in dramatically reducing
their rate of population growth. Although Saudi Arabia has
dramatically decreased its population growth rate, from 5.2
percent per year in the 1980s to 3.2 percent per year in the
1990s,xxxix such a rate of population growth leads to a doubling
of population in less than 25 years. This rate of growth has
produced a large youth cohort, combined with rapid expansion
of the labor force and rapid urbanization (urban growth of 7
percent per year in the 1980s and 4 percent per year in the
1990s xl ). The slow-down of the Saudi economy with the decline

FALL 2002 | 15
 Population and Security

in world oil prices portends poorly for absorbing this large

number of urban youth into the economy.
China has succeeded in cutting its overall population growth
and labor force growth to less than 1 percent per year. But
because of its enormous size, this still means finding new jobs
for roughly 13 million people per year. Far more important,
however, is the shift in Chinas population from the countryside
to the city. Because of the saturation of the agricultural sector,
population has been shifting to cities; virtually all of these new
job-seekers, plus many older agricultural workers, have been
pursuing urban employment. In an odd anomaly, despite very
low overall population growth, China has one of the worlds
fastest rates of urbanization, at nearly 5 percent per year in the
1980s and 4 percent per year in the 1990s. These rates,
combined with Chinas size, mean that in each decade,
approximately 150 million people have been added to the
population of Chinas cities and are dependent on urban jobs.
Until recently, Chinas enormous rate of economic growth,
averaging nearly 10 percent per annum, has allowed China to
absorb these job-seekers. Yet in the last year, Chinas economic
growth rate has dropped as the economy has tipped toward
deflation. A sustained collision between diminished economic
growth and the tens of millions moving to cities in search of
work every year bodes ill for social and political stability.
Therefore, while the marked decrease in population growth
in many countries and regions is good news for those concerned
about global population, it offers no clear relief for concerns
about the security implications of population change. Despite
slow-downs in overall growth, many countries may well
experience collisions between their agrarian populations and
access to land; between the expansion of their labor force,
educated aspiring elites, urban population and youth cohorts
and the absorption rate of their economies; and between
migrants and resident populations that inflame ethnic and
regional tensions.

FALL 2002 | 16
 Population and Security

DEMOGRAPHIC CHANGES AS INDICATORS AND OUTCOMES OF

VIOLENT POLITICAL CONFLICTS
Demographic factors have also proven highly useful in models
for forecasting political risks. In the work of the State Failure
Task Force, several demographic variables (including
urbanization to development ratio, life expectancy, adult and
infant mortality levels) were found to be useful predictors of
political violence, even after allowing for the impact of regime
type and such economic factors as international trade relations.xli
In particular, the rate of infant mortality was found to be an
important predictor of risk in almost all models, a result
confirmed by Urdal.xlii This is not because infant mortality itself
directly affects political processes. Instead, it appears that infant
mortality is the best single tool for assessing the wide variety of
factors (average income, income distribution, provision of health
care, nutrition) that affect the overall quality of life for
individuals in a society. High levels of infant mortality, relative
to world averages, indicate higher risks of political crises.
Nicholas Eberstadt has further argued that in communist
countries in particular, a rise in infant mortalitysomething
hardly ever seen, even in the Third Worldis a powerful
portent of coming upheaval.xliii Such a rise occurred in the Soviet
Union prior to its collapse and now appears to be occurring in
North Korea. These demographic changes may serve as a useful
early alert of coming security problems.
Finally, it should also be remembered that the relationship
between population changes and violent conflicts is not
unidirectional. Violent conflicts can also have large and long-
lasting impacts on demography. Revolutions frequently bring
marked shifts in marriage and birth rates (depending on whether
the post-revolutionary period is one of rampant optimism or
pessimism), in urbanization (if the new regime sponsors urban
development), in education (if the new regime dramatically
expands enrollments) and in migration (as the new regime and
the violence associated with it may either attract migrants from
abroad or send them across borders seeking escape from violence
or persecution). Violent conflicts rarely end conclusively; a more

FALL 2002 | 17
 Population and Security

common pattern is that cycles of violence succeed one another.

Part of the reason for this is that violent conflicts often produce
population changes that, in the next generation if not earlier,
feed back into the creation of renewed political risks.
For example, in Palestine the preservation of stateless Arabs
in refugee camps following the 1967 Israeli-Arab war led, twenty
years later, to the growth of a vast, aggrieved youth cohort with
limited economic prospects in the occupied territories. It was
this cohort that played a crucial role in the intifada uprisings in
Gaza and the West Bank. In Central Africa, the movement of
Tutsi and Hutu groups across borders as a result of internal
conflicts in Rwanda and Burundi led to destabilizing ethnic
conflicts in Congo-Zaire and to renewed and intensified conflicts
when new cohorts of formerly exiled Hutus and Tutsis returned
to their countries. Unless measures are taken to provide both
economic and political hope to the populations of present-day
Kosovo, Bosnia and Palestine, it is likely that the population
displacements that have occurred in those conflicts, combined
with the weak economic conditions and political institutions
facing the next cohort of young men growing up in those
regions, will produce not a lasting peace but a renewal of ethnic
conflicts. It thus appears that a focus on demographic changes
can be helpful both in alerting us to coming security problems
and in helping us foresee how these might fuel further problems
in the future.

Notes

i
United States Bureau of the Census, World Population Profile 1998. Report
WP/98 (Washington, DC: US Government Printing Office, 1999) p. 11.
ii
Ibid., p. 12
iii
Thomas Homer-Dixon, On the Threshold: Environmental Changes as
Causes of Acute Conflict, International Security 16 (1991) pp. 76116.
iv
Paul Diehl, Environmental Conflict: An Introduction, Journal of Peace
Research 35 (1998) pp. 275276.
v
Daniel Deudney, The Case Against Linking Environmental Degradation and
National Security, Millennium 19 (1990) pp. 461476. Marc A. Levy, Is the

FALL 2002 | 18
 Population and Security

Environment a National Security Issue? International Security 20, no. 2

(1995) pp. 3562.
vi
Nils Petter Gleditsch, Armed Conflict and the Environment, Journal of
Peace Research 35 (1998) pp. 381400.
vii
Wenche Hauge and Tanja Ellingsen, Causal Pathways to Conflict, Journal
of Peace Research 35 (1998) p. 314.
viii
Ibid., p. 311, Table II.
ix
Gnther Baechler, Why Environmental Transformation Causes Violence: A
Synthesis, Environmental Change and Security Project Report of the
Woodrow Wilson Center 4 (1998) p. 32; emphasis in original.
x
Daniel Esty, Jack A. Goldstone, Ted Robert Gurr, Barbara Harff, Pamela
Surko and Alan N. Unger, Working Papers: State Failure Task Force Report
and Daniel Esty, Jack A. Goldstone, Ted Robert Gurr, Barbara Harff, Marc
Levy, Geoffrey D. Dabelko, Pamela Surko and Alan N. Unger, State Failure
Task Force Report: Phase II Findings (McLean, VA: Science Applications
International Corporation, 1995 and 1998).
xi
Richard Hill, Swarupa Ganguli and Dede Naylor, Environmental
Flashpoints in South Asia, in Robert S. Chen, W. Christopher Lenhardt and
Kara F. Alkire, eds., Consequences of Environmental ChangePolitical,
Economic, Social; Proceedings of the Environmental Flash Points Workshop,
Reston, Virginia, 1214 Nov. 1997 (University Center, MI: Consortium for
International Earth Science Information Network, 1998) pp. 127176.
xii
Miriam R. Lowi, Water and Power: The Politics of a Scarce Resource in the
Jordan River Basin (Cambridge: Cambridge University Press, 1993).
xiii
Ronnie D. Lipschutz, Damming Troubled Waters: Conflict over the
Danube, 19502000. Paper presented at Environment and Security
Conference, Institute of War and Peace Studies, Columbia University, New
York, NY, 24 October 1997.
xiv
Peter H. Gleick, The Worlds Water: The Biennial Report on Fresh Water
Resources. (Washington, DC and Covelo, CA: Island. 1998).
xv
A recent article providing an excellent survey of water and security issues
notes that more than 3,600 treaties have been signed over different aspects of
international waters, with remarkable elegance and creativity in dealing with
water issues; moreover, in the last 3,000 years, there has never been a war
fought over water (Aaron T. Wolf, Water and Human Security, Aviso: An
Information Bulletin on Global Environmental Change and Human Security 3
[1999] p. 2). Miriam Lowi (Water and Conflict in the Middle East and South
Asia: Are Environmental Issues and Security Issues Linked? Journal of
Environment and Development 8 [1999] p. 389) similarly states that we have
not found, to date, cases in which interstate war derives primarily and
predominantly from the depletion, degradation, or inequitable distribution of
environmental resources such as freshwater. See also Aaron T. Wolf, Water
Wars and Water Reality, in Steve Lonergan, ed., Environmental Change,
Adaptation, and Human Security. Dordrecht: Kluwer (1999).

FALL 2002 | 19
 Population and Security

xvi
Ted Robert Gurr, Handbook of Political Conflict (New York: Free Press,
1980). Jack A. Goldstone, ed., The Encyclopedia of Political Revolutions
(Washington, DC: Congressional Quarterly, 1998) and Revolutions:
Theoretical, Comparative, Historical, and Historical Studies, 3rd ed. (Ft.
Worth: Wadsworth, 2002).
xvii
John Foran, ed., Theorizing Revolutions (London: Routledge, 1997); and
Goldstone, ed. (1998).
xviii
Valerie Percival and Thomas Homer-Dixon, The Case of South Africa,
Journal of Peace Research 35 (1998) pp. 279298; and Colin Kahl,
Population Growth, Environmental Degradation, and State-Sponsored
Violence: The Case of Kenya, 19911993, International Security 23 (1998)
pp. 80119.
xix
Alex de Sherbinen, World Population Growth and US National Security,
Environmental Change and Security Project Report of the Woodrow Wilson
Center 1 (1995) pp. 2429.
xx
Joseph Whitmeyer and Rosemary L. Hopcroft, Community, Capitalism, and
Rebellion in Chiapas, Sociological Perspectives 39, no. 4 (1996) pp. 517539.
xxi
Goldstone (1991); Goldstone, ed. (2002).
xxii
Ellen Brennan, Population, Urbanization, Environment, and Security: A
Summary of the Issues, Comparative Urban Studies Occasional Paper Series
no. 22 (Washington, DC: Woodrow Wilson International Center for Scholars,
1999).
xxiii
Esty et al. (1998) p. 15.
xxiv
William Doyle, The Price of Offices in Pre-Revolutionary France,
Historical Journal 27 (1984) pp. 831860. Lenore OBoyle, The Problem of
an Excess of Educated Men in Western Europe, 18001850, Journal of
Modern History 42 (1970) pp. 471495. Goldstone (1991) and The Soviet
Union: Revolution and Transformation, in Mattei Dogan and John Higley,
eds., Elites, Crises, and the Origins of Regimes (Lanham, MD: Rowman and
Littlefield, 1998) pp. 95124.
xxv
Samuel P. Huntington, The Clash of Civilizations, and the Remaking of
World Order (New York: Simon and Schuster, 1996).
xxvi
Paul Collier, Doing Well Out of War: An Economic Perspective, in Mats
Berdal and David M. Malone, eds., Greed and Grievance (Boulder, CO: Lynne
Rienner, 2000) pp. 91111.
xxvii
Herbert Moller, Youth as a Force in the Modern World, Comparative
Studies in Society and History 10 (1968) pp. 238260; Goldstone (1991).
xxviii
Christian G. Mesquida and Neil I. Weiner, Male Age Composition and
Severity of Conflicts, Politics and the Life Sciences 18 (1999) pp. 113117.
xxix
Henrik Urdal, Population Pressure and Domestic Conflict: Assessing the
Role of Youth Bulges in the Onset of Conflict 19502000; paper presented
at the Fourth Pan-European International Relations Conference, University of
Kent (Canterbury, 810 Sept. 2001).

FALL 2002 | 20
 Population and Security

xxx
David Laitin and James Fearon, Ethnicity, Insurgency, and Civil War,
American Political Science Review (Forthcoming); Nicholas Sambanis, Do
Ethnic and Nonethnic Civil Wars Have the Same Causes? Journal of Conflict
Resolution 45 (2001) pp. 259282; Steven C. Poe and C. Neal Tate,
Repression of Human Rights to Personal Integrity in the 1980s: A Global
Analysis, American Political Science Review 88 (1994) pp. 853872; Poe,
Tate and L.C. Keith, Repression of the Human Right to Personal Integrity
Revisited: A Global Cross-national Study Covering the Years 19761993,
International Studies Quarterly 43 (1999) pp. 291313.
xxxi
Conway Henderson, Population Pressures and Political Repression, Social
Science Quarterly 74 (1993) pp. 322333.
xxxii
Nathaniel Beck, Time-SeriesCross-Section Data: What Have We
Learned in the Past Few Years, Annual Review of Political Science 4 (2001)
pp. 271293; Beck, Jonathan N. Katz and Richard Tucker, Taking Time
Seriously: Time-SeriesCross Section Analysis with a Binary Dependent
Variable, American Journal of Political Science 42 (1998) pp. 12601288.
Kristian S. Gleditsch and Michael D. Ward, War and Peace in Space and
Time: The Rold of Democratization, International Studies Quarterly (2000)
pp. 129; Goldstone, The Case for Case Control: Substantive Issues in the
Study of Rare and Calamitous Events (manuscript, University of California,
Davis, 2002); Donald Green, Soo Yeon Kim, and David Yoon, Dirty Pool,
International Organization 55 (2001) pp. 441468; Gary King and Langche
Zeng, Logistic Regression in Rare Events Data, Political Analysis 9 (2001)
pp. 135163.
xxxiii
Esty et al (1998).
xxxiv
Beck, Gary King and Langche Zeng, Improving Quantitative Studies of
International Conflict: A Conjecture, American Political Science Review 94
(2000) pp. 2135.
xxxv
Charles Ragin, Fuzzy-Set Social Science (Chicago: University of Chicago
Press, 2000).
xxxvi
Myron Weiner and Sharon Stanton Russell, Demography and National
Security (Providence, RI: Berghahn, 2000).
xxxvii
Michael S. Teitelbaum and Jay Winters, A Question of Numbers: High
Migration, Low Fertility, and the Politics of National Identity (New York: Hill
and Wang/Farrar, Straus & Giroux, 1998).
xxxviii
David Lane, The Gorbachev Revolution: The Role of the Political Elite in
Regime Disintegration, Political Studies 44 (1996) pp. 423; Michael Urban,
Vyacheslav Igrunov and Sergei Mitrokhin, The Rebirth of Politics in Russia
(Cambridge: Cambridge University Press, 1997); Michael McFaul, Russias
Unfinished Revolution: Political Change from Gorbachev to Putin (Ithaca, NY:
Cornell University Press, 2001).
xxxix
World Bank, World Development Indicators 1997 (CD-ROM, 1998).
xl
Ibid.
xli
Esty et al. (1995, 1998).

FALL 2002 | 21
 Population and Security

xlii
Urdal (2001).
xliii
Nicholas Eberstadt, The Poverty of Communism (New Brunswick, NJ:
Transaction Books, 1988).

FALL 2002 | 22
The author has requested enhancement of the downloaded file. All in-text references underlined in blue are